Build Better Generative Adversarial Networks (GANs)

Monthly Subscription Starting at AED 99 + VAT

Level Professional
Duration 29 hours
Course by DeepLearning.AI
Offered by

About

In this course, you will: - Assess the challenges of evaluating GANs and compare different generative models - Use the Fréchet Inception Distance (FID) method to evaluate the fidelity and diversity of GANs - Identify sources of bias and the ways to detect it in GANs - Learn and implement the techniques associated with the state-of-the-art StyleGANs The DeepLearning.AI Generative Adversarial Networks (GANs) Specialization provides an exciting introduction to image generation with GANs, charting a path from foundational concepts to advanced techniques through an easy-to-understand approach. It also covers social implications, including bias in ML and the ways to detect it, privacy preservation, and more. Build a comprehensive knowledge base and gain hands-on experience in GANs. Train your own model using PyTorch, use it to create images, and evaluate a variety of advanced GANs. This Specialization provides an accessible pathway for all levels of learners looking to break into the GANs space or apply GANs to their own projects, even without prior familiarity with advanced math and machine learning research.

Modules

Evaluation of GANs

10 Videos

8 Readings

1 Lab

1 Programming

Show info about module content

1 Programming

Fréchet Inception Distance

1 Labs

(Optional) Perceptual Path Length

10 Videos

Welcome to Course 2
Welcome to Week 1
Evaluation
Comparing Images
Feature Extraction
Inception-v3 and Embeddings
Fréchet Inception Distance (FID)
Inception Score
Sampling and Truncation
Precision and Recall

8 Readings

Syllabus
[IMPORTANT] Have questions, issues or ideas? Join our Forum!
(Optional) A Closer Look at Inception Score
(Optional) HYPE!!
(Optional) More on Precision and Recall
(Optional) Lecture Notes W1
(Optional) Recap of FID and IS
Works Cited

GAN Disadvantages and Bias

6 Videos

9 Readings

1 Lab

1 Programming

1 Assignment

Show info about module content

1 Programming

Bias

1 Assignment

Analyzing Bias

1 Labs

Alternatives: Variational Autoencoders (VAEs)

6 Videos

Welcome to Week 2
Disadvantages of GANs
Alternatives to GANs
Intro to Machine Bias
Defining Fairness
Ways Bias is Introduced

9 Readings

(Optional Notebook) Score-based Generative Modeling
Machine Bias
Fairness Definitions
A Survey on Bias and Fairness in Machine Learning
Finding Bias
(Optional) Lecture Notes W2
(Optional Notebook) GAN Debiasing
Works Cited
(Optional Notebook) NeRF: Neural Radiance Fields

StyleGAN and Advancements

9 Videos

6 Readings

2 Lab

1 Programming

Show info about module content

1 Programming

Components of StyleGAN

2 Labs

(Optional) Components of StyleGAN2
(Optional) Components of BigGAN

9 Videos

Welcome to Week 3
GAN Improvements
StyleGAN Overview
Progressive Growing
Noise Mapping Network
Adaptive Instance Normalization (AdaIN)
Style and Stochastic Variation
Putting It All Together
Conclusion of Course 2

6 Readings

(Optional) Lecture Notes W3
[IMPORTANT] Reminder about end of access to Lab Notebooks
(Optional) The StyleGAN Paper
(Optional) StyleGAN Walkthrough and Beyond
Works Cited
Acknowledgments

Auto Summary

Unlock the potential of Generative Adversarial Networks (GANs) with this comprehensive course in Data Science & AI. Led by DeepLearning.AI on Coursera, you'll explore GAN challenges, evaluate models using the Fréchet Inception Distance (FID) method, and tackle bias detection. Dive into state-of-the-art StyleGANs, learn to train models with PyTorch, and generate images. This 1740-minute Professional level course is perfect for learners of all levels, offering hands-on experience and insights into GANs' social implications. Available with a Starter subscription, it's ideal for those looking to advance their AI skills.